SortedRL: Accelerating RL Training for LLMs through Online Length-Aware Scheduling
This addresses efficiency issues in RL training for large language models, particularly for tasks requiring long chain-of-thought generation, though it appears incremental as it optimizes an existing pipeline.
The paper tackled the bottleneck in RL training for LLMs, where rollout phases consume up to 70% of training time for long trajectories, by proposing SortedRL, an online length-aware scheduling strategy that reduces RL training bubble ratios by over 50% and achieves performance gains of 3.9% to 18.4% over baselines.
Scaling reinforcement learning (RL) has shown strong promise for enhancing the reasoning abilities of large language models (LLMs), particularly in tasks requiring long chain-of-thought generation. However, RL training efficiency is often bottlenecked by the rollout phase, which can account for up to 70% of total training time when generating long trajectories (e.g., 16k tokens), due to slow autoregressive generation and synchronization overhead between rollout and policy updates. We propose SortedRL, an online length-aware scheduling strategy designed to address this bottleneck by improving rollout efficiency and maintaining training stability. SortedRL reorders rollout samples based on output lengths, prioritizing short samples forming groups for early updates. This enables large rollout batches, flexible update batches, and near on-policy micro-curriculum construction simultaneously. To further accelerate the pipeline, SortedRL incorporates a mechanism to control the degree of off-policy training through a cache-based mechanism, and is supported by a dedicated RL infrastructure that manages rollout and update via a stateful controller and rollout buffer. Experiments using LLaMA-3.1-8B and Qwen-2.5-32B on diverse tasks, including logical puzzles, and math challenges like AIME 24, Math 500, and Minerval, show that SortedRL reduces RL training bubble ratios by over 50%, while attaining 3.9% to 18.4% superior performance over baseline given same amount of data.